In South Africa (SA), an estimated 51,300 children between 1 month and 5 years old die every year. -- the majority (57%) due to HIV. Childhood malnutrition, associated with diminished immune function, is a major contributor to these deaths. Key interventions to reduce HIV- and nutrition- related child mortality include well- defined and evidence-based packages of care to improve survival of HIV-infected adult caregivers as well as HIV-exposed children. Community health workers (CHW), if appropriately supported, can play a critical role in improving access to care and supporting delivery of evidence-based HIV and maternal and child health and nutritional (MCHN) interventions. While prior studies have shown the feasibility of CHW-led models for MCHN health promotion and their impact on some maternal-child health outcomes, evaluations of non-HIV focused CHW programs have shown CHW adherence to protocols to decline rapidly after training. Thus, despite the recognized importance of CHW models, there is little research to address the effectiveness of strategies for training CHW or for ensuring quality of care and fidelity to evidence-based protocols. We propose a 3-year study to develop and test an mHealth innovation to support CHW in their delivery of home-based HIV and MCHN education, support, and referral in a high HIV prevalence community in KwaZulu Natal, SA. Community Health Worker Assistive Technologies (CHAT) will be developed on a handheld platform (i.e., a multi-function tablet computer) and will leverage available, low-cost devices and open source software. CHAT will provide CHW with sustained, integrated support including protocol training, powerful handheld technologies and accessible, media enhanced resources (e.g., health education videos and animations) that will promote interactive discussions with clients. It will enable electronic data capture that supports high quality information gathering using data validation and branching logic, and will support communication with clinic-based health providers and CHW supervisors. To test the impact of CHAT, we will conduct a small community randomized controlled trial (C-RCT) to investigate the impact of CHAT on 1) CHW quality of care as well as knowledge about HIV, nutrition, danger signs of childhood illness, and child development; 2) health care utilization of caregivers for children, and child-care-related behaviors (e.g., immunizations, hygiene practices); and 3) the preliminary impact of CHAT on child health including growth, disease, and development for HIV-affected children age 0 to 59 months old. Participants will include 60 CHW (30 intervention, 30 control) and 80 caregiver-child pairs (40 intervention and 40 control) who will be followed for one year and assessed at regular intervals for the outcomes of interest. We will also assess the acceptability and feasibility of CHAT, conducting health and developmental assessments at children's homes, and translate lessons learned in the development study into the design of a large C-RCT. PUBLIC HEALTH RELEVANCE: Community health workers (CHW) can play a critical role in improving access to and continuity of care and treatment for underserved populations. This project will develop and assess the preliminary impact of an mHealth intervention targeting CHW who provide home-based support for HIV-affected households in KwaZulu Natal Province, South Africa on 1) CHW knowledge about HIV, maternal and child health and nutrition, and child development and on quality of care; 2) caregiver health behaviors and health care utilization for HIV- affected children ages 0-59 months old; and 3) the preliminary impact of the intervention on child growth, disease and development of HIV-affected children ages 0-59 months old. Ultimately, the project has the potential to improve the ways in which community-based health care and prevention services are conceptualized and delivered throughout African communities by demonstrating the role of mHealth in improving CHW practice and improving the health of HIV-affected children and their caregivers.